Questionnaire Design and Cleaning Data
How to gather good data: designing clear, unbiased questionnaire questions, and cleaning raw data to fix errors, missing values and other problems before analysis.
Get the method right under pressure
Free interactive practice on the steps that lose marks under exam pressure.
Start revising freeWhat you'll cover
Good questions, clean data
Good data starts with good questions. This module looks at how to design a questionnaire, from open and closed questions to avoiding leading questions, and at how to clean data by fixing errors, missing values and other problems before it is analysed.
Words for data collection
Learn these four terms before you design a questionnaire.
Open and closed
The two main question types have different strengths.
Say why the design matters
To reach the top grades, explain why a design choice matters, for example that a leading question biases the results, and that overlapping response groups make a value impossible to place.
Match the term to its meaning
- closed question
- open question
- leading question
- biased source
- a question with fixed answer options
- a question with a free written answer
- a question that pushes you toward an answer
- a source that gives an unfair, one-sided view
Match the data problem
- missing data
- non-response
- incorrect format
- incomplete response
- a value that was not recorded
- when someone does not answer at all
- data written in the wrong form, like a date as text
- an answer that is only partly filled in
What is a leading question?
What is a leading question?
- A question that pushes the respondent toward a particular answer.
- A question with a free written answer.
- The first question on a form.
- A question with a fixed list of answers to tick.
Features of a good question
Select the TWO features of a good question.
- It is clear and simple.
- Its answer options do not overlap.
- It leads the respondent to an answer.
- It uses a biased source.
Order the survey
Put the steps of a survey in order, earliest first.
- Design the questionnaire
- Collect the data from respondents
- Clean the data to fix problems
- Analyse the cleaned data
Summarise the ideas
A _____ question has fixed answer options, while an _____ question has a free written answer. A _____ question should be avoided because it pushes people toward an answer. Response options must not _____. Before analysis, raw data is _____ to fix errors and missing values.
Count the groups
In a good questionnaire, each answer should fit into how many response groups?
Spot the two data problems
Tap the TWO data problems that cleaning fixes.
- missing data
- non-responses
- a clear question
- a random sample
- a fair source
Good, leading or overlapping?
Decide what is wrong, or right, with each question.
- How old are you? 0 to 10, 11 to 20, 21 to 30
- Do you agree that our great product is the best?
- How old are you? 0 to 10, 10 to 20, 20 to 30
Design a good question
Explain how to design a good questionnaire question. Describe two features a good question should have, and one thing it should avoid.
- Describe one feature of a good question
- Describe a second feature
- Describe one thing a good question should avoid